RE: minor corrections

Hi Mark,

Thanks for the rapid feedback!

> -----Original Message-----
> From: Mark Jones [mailto:jones@research.att.com]
> Sent: 27 March 2001 17:11
> To: skw@hplb.hpl.hp.com
> Cc: jones@research.att.com
> Subject: minor corrections
> 
> 
> Stuart,
> 
> I just went through the March 27 version of the AM and picked up
> a few corrections.
> 
> 1. The prose near the beginning of section 2 mentions XMLP_DATA
> as an operation (which is now gone from Fig 2.1).

Thanks. I've now corrected my working copy.

> 
> 2. Figure 3.1's title should probably be XMLP_UNITDATA Operation.

Yes... thanks, again corrected in my working copy.

> 3. In Section 3.3, Operation Parameters, it lists Message.Headers and
> Message.Bodies.  The message processing model does not distinguish
> between header and body blocks.  One of the reasons why I thought it
> was good to eliminate the distinction if we didn't otherwise need it
> is to make it easier to serialize the contents of a reply message.
> Multiple handlers at the receiver may be invoked (on various blocks).
> They may each produce blocks to contribute in the response.  
> In the SOAP scheme, the response blocks would have to be grouped as
entries in the
> correct construct (SOAP:Header or SOAP:Body).  If we simply have
> undifferentiated XMLP:Block, then it may be possible for the processor
> to compose the reply message on the fly as it serially executes
> handlers.  This may simply buffering.
> 
> How about just listing Message.Blocks in 3.3?

That would be although I would like to feel there was a broad concensus
about the absense of a distinction between headers and bodies. I'll hold any
changes pending discussion.

> 
> --mark
> 
> 
> Mark Jones
> AT&T Labs

Many thanks

Stuart
PS. I've Cc' xml-dist-app so that folks know the 'bugs' have been picked up
and to encourage some discussion of your point 3.

Received on Tuesday, 27 March 2001 11:28:20 UTC